This application is the A1 resubmission of R01 AG033575. Among older adults, both Major Depressive Disorder (MDD) and chronic low back pain (CLBP) are common, frequently comorbid, and associated with substantial disability. Treatment with combination pharmacotherapy and psychotherapy is often required for patients with difficult to treat MDD and CLBP, especially when they are comorbid. The primary question addressed by this 2-phase RCT is to answer: What is the value of combination treatment with high-dose duloxetine (DUL) plus a symptom-specific and age-appropriate intervention called Problem Solving Therapy for Depression and Pain (PST-DP) for older adults living with MDD and CLBP when primary pharmacotherapy with low-dose duloxetine and supportive management (DUL/SM) has led to partial or non-response? Step 1 is 8 weeks and consists of open treatment with low-dose (90 mg/day) DUL/SM. Response is defined as a Patient Health Questionnaire score of < 5 and at least a 30% improvement on the Pain Numeric Rating Scale. Step 1 non-responders progress to step 2, a randomized controlled trial of comparing 1) high-dose DUL/SM (120 mg/day) and 2) high-dose DUL/PST-DP (120 mg/day). Step 2 is 10 visits over 14 weeks. DUL/PST-DP treats MDD and CLBP by: 1) building upon antidepressant gains achieved with low dose DUL/SM, 2) improving self- efficacy to reduce learned helplessness, and 3) activating participants via activity scheduling. DUL/SM involves: 1) education about the chronicity of MDD and CLBP, 2) an emphasis on adherence, and 3) management of side effects and suicide risk. This study will be implemented in primary care, older adults' preferred site for depression treatment. We will recruit 250 subjects into step 1. 150 non-responders will be randomized in step 2 to detect an effect size of at least 0.30 between groups. Our primary hypotheses are that more subjects randomized to DUL/PST-DP will respond and experience less back-related disability compared to subjects randomized to DUL/SM. Survival analysis will be used to compare response rates and time to response for the treatment conditions. In addition to testing an empirically-developed combination treatment for low dose DUL/SM non-responders, this application by a new investigator is novel because it is testing DUL/PST-DP in primary care, the site where most patients with MDD and CLBP are seen. Improvement in both depression and pain intensity as criteria for response is novel and adds to the public health significance and innovation. After completion of step 2, all treatment responders will be assessed for 12 months, to compare treatment durability. Our proposal responds to NIA PA-07-282, Mechanisms, Models, Measurement, & Management in Pain Research by engaging an interdisciplinary team that utilizes both pharmacologic and nonpharmacologic approaches to pain management in older adults. Multidisciplinary approaches are required for complex pain conditions. However, we are unaware of findings from any program or stepped care trial that addresses DSM-IV-defined MDD and CLBP in primary care older adults simultaneously.